Respiratory illness and diarrhoea continue to be the leading causes of paediatric morbidity and mortality in the Dominican Republic. An important first step in alleviating this disease burden is to understand patterns and predictors of health services utilization for these conditions. This study examines the predictors of (a) health services utilization, and (b) public versus private sector use, for respiratory illness in the under-five population in the Dominican Republic. The DHS-2 dataset (1991) was utilized for analysis. Logistic regression models for predicting use and non-use, and for predicting private versus public sector use, were constructed using the Andersen Behavioural Model as the conceptual framework. Our findings indicate that sex, location and possession index quartile are factors that influence the decision to seek care or not for respiratory illness in under-fives. In contrast, the choice between the public and private sector is determined by location and insurance status. From the policy perspective, if the Dominican Republic were to undertake steps to increase private insurance coverage, our results indicate that this would lead to increased utilization of private sector providers for respiratory illness by children having private insurance, but would not have an impact on overall utilization (i.e. use vs. non-use). On the other hand, one of the ways to deliver cost-effective interventions by the publicly financed system would be to improve facilities in the rural areas.